I can't find info on Google Doc's that mentioned it. I found this that says:
If you want to simply remove sensitive data from your input content, Cloud DLP supports a redaction transformation (RedactConfig in the DLP API).
Source: Google Doc
Does remove mean delete and irreversible?
Redaction deletes the part of the data that is specified to remove and it is not reversible.
Choosing the de-identification transformation depends on the data you want to de-identify and the purpose of de-identifying the data.
Cloud DLP supports some de-identification techniques and in that
Redaction is used to delete all or part of a detected sensitive value.If you want to simply remove sensitive data from the input, Cloud DLP supports a redaction transformation (RedactConfig in the DLP API).
redactConfig
Specifying redactConfig redacts a given value by removing it completely. The redactConfig message has no arguments; specifying it enables its transformation.
Crypto-based tokenization encrypts the original sensitive data using a cryptographic key. Cloud DLP supports several types of tokenization, including transformations that can be reversed, or "re-identified."
Reversible transformations can be reversed to re-identify the sensitive data using the content.reidentify method.
Refer to the link for more information on Crypto-based tokenization.